Fix mutex locking bugs and improve performance. Fix interval no longer
set statically since it is available via `set_interval` uORB function.
Activation/deactivation no longer triggers standby/hotstart since this
functionality was very unstable.
Signed-off-by: Matteo Golin <matteo.golin@gmail.com>
Describes some of the quirks of using littlefs on NuttX and links to an
issue discussing the sync semantics.
Signed-off-by: Matteo Golin <matteo.golin@gmail.com>
Without c++ standard includes, the <atomic> library is not available.
GCC may use __auto_type for the C-style atomic oprations, and this
is not available for C++. Just define the __auto_type to auto for C++.
Also, the built-in _atomic functions for GCC expect "volatile" parameters
in C, but don't allow that in C++. So selecting the atomic_t and atomic64_t
types according to that.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
This commit introduces a timer capture driver for the STM32H7 series
ported from the STM32 F series.
The main changes include:
- A new generic timer capture driver for STM32H7.
- A lower-half driver to integrate with the NuttX capture subsystem.
- Kconfig options to enable and configure capture for various timers.
- Pin definitions for TIM1-4 capture inputs on the nucleo-h743zi.
- An update to `cap_register_multiple` to handle multiple device registration.
- An update to `stm32_bringup` to register the capture drivers.
The current implementation is based on a driver originally for PWM input,
and as such, it calculates duty cycle and frequency. It is also limited
to a single capture channel per timer.
The original implementation's `stm32_cap_init` in
`arch/arm/src/stm32h7/stm32_capture.c` has been modified to accept a
channel number instead of using a hardcoded 0 through
`STM32_CAP_CHANNEL_COUNTER`.
This serves as a foundation for future development of more comprehensive
input capture capabilities on STM32H7 platforms.
Tested by polling and reading `/dev/cap0-4` with
`ioctl(fds[i], CAPIOC_FREQUENCE, freq)` while sending a square wave signal to
appropriate pins and checking frequency.
Also tested by bypassing upper half driver and setting up capture on
TIM4 channels 1-4 as explained in #16762.
Signed-off-by: Côme VINCENT <44554692+comejv@users.noreply.github.com>
The documentation for using the Unity testing framework previously just
linked to the official Unity website. In using Unity, I noticed there
were much more extensive docs on the actual GitHub page. This PR
includes references to those docs and also provides a short-winded
explanation of how to make a Unity test suite on NuttX.
Signed-off-by: Matteo Golin <matteo.golin@gmail.com>
The GNSS lower-half driver for facilitating the implementation of uORB
GPS/GNSS device drivers was lacking documentation. Contributors cannot
be expected to implement their drivers correctly with this framework if
it is undocumented. This change adds documentation for how to use the
lower-half GNSS driver, provides an example and also adds a notice for
users to use this framework if implementing their own GNSS driver.
Signed-off-by: Matteo Golin <matteo.golin@gmail.com>
This commit simplifies sending errors back to OP-TEE and avoids
code duplication when handling errno values.
Signed-off-by: Theodore Karatapanis <tkaratapanis@census-labs.com>
This commit expands RPC support for the OP-TEE driver using 2 files:
1) drivers/misc/optee_rpc.c
* Add support for RPCs that can be handled directly by the kernel.
* Can delegate RPC handling to optee_supplicant.c for RPCs that
need userspace interaction.
2) drivers/misc/optee_supplicant.c
* Enable communication between the userspace TEE supplicant and the
kernel driver.
Additional changes were needed to the following files:
1) drivers/misc/optee.c
* Add ioctls used SOLELY by the userspace TEE supplicant.
* Register /dev/teepriv0 if the supplicant is enabled in Kconfig
* Add OPTEE_ROLE_CA and OPTEE_ROLE_SUPPLICANT conditionals to
differentiate paths, between a normal Client Application (CA)
and the TEE supplicant.
* Change some functions from static to "public" to reuse them
in other C files.
* Adjust optee_to/from_msg_param() to work with RPCs.
2) drivers/misc/optee_smc.c
* Call the RPC handler from optee_rpc.c
3) drivers/misc/optee_msg.h
* Add definition needed for RPCs
4) drivers/misc/tee.h
* Add ioctl definitions
* Add TEE_SHM_SUPP flag, checked when unregistering supplicant
memory.
5) Documentation/guides/optee.rs
* Add documentation for RPCs and the supplicant.
6) drivers/misc/{CMakeLists.txt, Make.defs}
* Account for the new files.
7) drivers/misc/Kconfig
* Add DEV_OPTEE_SUPPLICANT option to enable/disable the supplicant
driver.
Signed-off-by: Theodore Karatapanis <tkaratapanis@census-labs.com>
The previous approach with memfd has 3 problems:
1) The close operation on the memfd isn't tied with optee_shm_close,
therefore close(fd) doesn't free the optee_shm struct allocated
by the kernel.
2) The kernel unnecessarily maps the file descriptor to its memory,
however only userspace should need to do that.
3) Since the kernel doesn't need to map the file descriptor we
don't need to unmap it.
To use anonymous mapping, the prototype of map_anonymous() was
moved from fs/mmap/fs_anonmap.h to include/nuttx/fs/fs.h. Since
fs_anonmap.h didn't contain any other information it is deleted.
A type from fs/mmap/fs_rammap.h was moved to the public :
include/nuttx/fs/fs.h as well.
Signed-off-by: Theodore Karatapanis <tkaratapanis@census-labs.com>
The VA -> PA translation must be performed once, at allocation time,
while the caller's virtual address space is known. If a second process
later tries to translate the same VA from a different mapping, the
calculated physical address can be wrong.
Signed-off-by: Theodore Karatapanis <tkaratapanis@census-labs.com>
1. Adopt an exponential backoff strategy () to dynamically adjust the timeout
duration, which is suitable for retry scenarios.
2. Optimize the default configuration to set the initial timeout to 5 seconds.
3. Support dynamic modification of the maximum timeout limit to adapt
to different scenarios.
Reference: RFC 1536 (section on retransmission implementation recommendations)
Signed-off-by: nuttxs <zhaoqing.zhang@sony.com>
NVBLK provides a block device that operates on top of a non volatile
memory (as a mtd device) that enables wear levelling for non volatile
memory. It's operation is similar to the dhara wear levelling library,
but nvblk is meant to be used on smaller (nor, mram, rram) memory.
I am also the author and maintainer of the nvblk library.
A block device can be created during startup by using:
```
nvblk_initialize(0, mtd, CONFIG_MTD_NVBLK_DEFAULT_LBS,
CONFIG_MTD_NVBLK_DEFAULT_IOBS,
CONFIG_MTD_NVBLK_DEFAULT_SPEB);
```
and a fat filesystem on top of this as:
```
nsh> mkfatfs /dev/mtdblock0
nsh> mount -t vfat /dev/mtdblock0 /mnt
```
this fat filesystem can then be used:
```
nsh> echo "test" > /mnt/test.txt
nsh> cat test.txt
test
nsh>
```
Signed-off-by: Laczen JMS <laczenjms@gmail.com>
This fixes the same issue for other targets, which was already fixed for
xtensa in commit 50d94863.
After the signals have been delivered, the local irqs need to be
disabled until the context switch. But just calling
leave_critical_section(regs[xx]) will enable them if they were
enabled in the context.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Mostly for CI purposes to validate OP-TEE changes
against a flat build. It has not been runtime-tested
with any specific setup involving an actual OP-TEE.
Signed-off-by: George Poulios <gpoulios@census-labs.com>
Previous change that introduced yielding upon OP-TEE
return with reason foreign interrupt, omitted to include
<sched.h>. This worked on kernel builds, but not on flat
builds. Include it.
Signed-off-by: George Poulios <gpoulios@census-labs.com>
Clear dma interrupt flags as before.
Change HTEF to HTF, fixes typo.
Remove ALL_FLAGS define. Clear CXFCR with all 1s like other instances do.
Handle half-transfer interrupt only if in circular mode.
Add ADC_HAVE_DMA guard around adc_rstart_dma. It is needed because priv->circular is guarded by ADC_HAVE_DMA.
After the signals have been delivered, the local irqs need to be
disabled until the context switch. But just calling
leave_critical_section(regs[REG_PS]) will enable them if they were
enabled in the context.
Any interrupt on this cpu before xtensa_context_restore() would set
the rtcb->xcp.regs to NULL, causing a crash in context restore.
Fix this by calling leave_critical_section(up_irq_save()) instead.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
This reverts commit 8f96e59aee.
The implementation of __dso_handle is independent of the HOST OS but dependent on the specific compiler.
Therefore, this patch will cause compilation failures for other cross-compilers on the Windows platform, such as arm-none-eabi-gcc and aarch64-none-elf-gcc.
Constraints should be applied to compiler target rather than compiler host.
CPU heavy OP-TEE calls would starve the Normal World tasks.
OP-TEE foreign interrupts were delivered to NW but we would
resume OP-TEE immediately without giving any potentially
unblocked tasks a chance to run. This would violate real-
time guarantees. Fix that by calling sched_yield() during
OP-TEE return with reason foreign interrupts.
Signed-off-by: George Poulios <gpoulios@census-labs.com>
Many iMXRT MCUs have internally two USB OTG controllers, however
NuttX currently only supports one USB controllers. This patch will
prepare the "house" to support both ports at same time.
Signed-off-by: Alan C. Assis <acassis@gmail.com>
* Recurrency is removed from filesystem directory rename.
* Fixes use after free in buffer that was used as output and argument.
Signed-off-by: Tomasz 'CeDeROM' CEDRO <tomek@cedro.info>
PIC32CZ CA70 family is pin to pin and binary compatible with
SAMV70/SAME70 families, therefore the support is placed in samv7
directory. The only difference is larger RAM memory compared to SAM
families.
Signed-off-by: Michal Lenc <michallenc@seznam.cz>
fb_notify_vsync is called when TE irq comming, and fb_remove_paninfo is called after paninfo comsumed
Signed-off-by: jianglianfang <jianglianfang@xiaomi.com>
